Early recognition of post-stroke spasticity: the I-REFER study

Background: Effective post-stroke spasticity (PSS) management relies on early recognition and treatment, but access to specialist physicians is limited and PSS remains underdiagnosed and undertreated. Empowering non-specialist nurses and therapists to identify potential harmful PSS may expedite appropriate referral. We assessed validity and inter-rater reliability (IRR) of the spastic movement disorders decision tree for upper extremity spasticity by comparing referral decisions made by nurses and therapists to those made by specialist physicians (https://clinicaltrials.gov/study/NCT06381999). Methods: = 69) by five nurses (primary objective) and four therapists (secondary objective) were compared with standard specialist physician decisions (two physicians). Results: IRR was high between nurses and physicians on the need for specialist referral, overall (85.5%) and at the elbow (88.4%), wrist (85.5%), and fingers (84.1%), with 84.6% sensitivity (to correctly identify patients requiring specialist referral) and 86.7% specificity (to identify those not requiring specialist referral). Positive and negative predictive values were also high overall (89.2 and 81.3%, respectively) and at all (movement) segments. Findings were similar for therapists versus physicians. Sensitivity analyses based on a categorical endpoint (low risk for PSS, assess botulinum toxin type A [BoNT-A] indication, indication for BoNT-A) showed similar trends, although IRR versus physicians was lower for both nurses and therapists compared with dichotomous results. Conclusion: This decision tree for spasticity in the upper extremities showed high validity and reliability for effectively helping nurses and therapists recognize potential PSS and initiate specialist referral for timely diagnosis and potential BoNT-A treatment. It is recommended that this tool be introduced into standard clinical practice and its potential for broader use in community settings be examined.